Reading-Assistance Tools Among Deaf and Hard-of-Hearing Computing Professionals in the U.S.: Their Reading Experiences, Interests and Perceptions of Social Accessibility
Oliver Alonzo, Lisa Elliot, Becca Dingman, Sooyeon Lee, Akhter Al Amin, Matt Huenerfauth · 2022 · ACM Transactions on Accessible Computing · doi:10.1145/3520198
Summary
This research investigates the reading experiences, needs, and interests of Deaf and Hard-of-Hearing (DHH) computing professionals in Automatic Text Simplification (ATS) tools—software that automatically rewrites complex text to make it easier to read. The study addresses a gap in prior ATS research, which has largely focused on evaluating simplification quality rather than understanding specific user groups' requirements in professional contexts. The researchers conducted a mixed-method study across two phases. Study 1 included pilot interviews (N=12), an online survey (N=32), and follow-up interviews (N=5) examining reading habits, experiences with complicated text, and interest in ATS tools. Study 2 comprised additional interviews (N=7) focused specifically on social accessibility—how perceptions of assistive technology use in workplace contexts might affect adoption. Participants were recruited from computing and IT fields, with diverse hearing identities (deaf, hard-of-hearing, culturally Deaf), communication preferences (spoken language, sign language, or mixed), and English comfort levels. The research is grounded in several important contexts: DHH adults show great diversity in literacy skills, computing professionals must constantly read to stay current, and DHH individuals are underrepresented in computing (only 0.8% of Stack Overflow respondents). Prior research on ATS has shown benefits for DHH readers, but no work had examined their specific interests in professional settings or the social implications of using such tools.
Key findings
DHH computing professionals read frequently—the majority reported reading at least weekly, with most reading occurring on electronic screens rather than paper. Primary reading purposes were work (N=29), academic (N=26), and personal communication (N=25). Participants reported reading more often than watching videos for learning about technical topics, though videos were preferred for unfamiliar or practical topics. When encountering complicated text, participants most commonly used dictionary lookups (N=25) or searched for alternative texts on the same topic (N=21). Unfamiliar vocabulary was identified as a primary source of difficulty, causing frustration, loss of confidence, and disruption to reading flow. Interest in ATS tools was strong: 25 of 32 survey respondents "somewhat agreed" or "strongly agreed" they would use such tools. The critical finding concerns social accessibility: participants wanted to be seen as competent, knowledgeable, and skilled in the workplace, but there is a perceived cultural relationship between reading skill and intelligence. Using ATS tools publicly could imply literacy difficulties, conflicting with desired professional image. Participants suggested mitigations: framing tools as "productivity tools" rather than assistive technology, making interfaces blend with other software, enabling quick hiding via keyboard shortcuts, and using alternative names like "translator," "paraphrase tool," or "efficient studying tool." Participants also noted these tools would benefit non-native English speakers and second-language learners.
Relevance
This research has significant implications for designing reading assistance technologies for professional contexts. The social accessibility findings reveal that functional benefits alone are insufficient—tools must be designed to avoid stigmatization and preserve users' professional image. For practitioners, the study provides specific design guidance: (1) give users control over how much text is simplified at once, (2) allow tools to blend in with existing software and be easily hidden, (3) consider accuracy concerns as users worried about meaning changes, (4) frame marketing as universal productivity tools rather than disability-specific assistive technology. The finding that participants preferred attempting to understand text independently before asking colleagues suggests ATS tools could reduce workplace barriers to seeking help. The research also highlights that reading assistance needs extend beyond DHH users to anyone for whom English is a second language—participants drew parallels between their experiences and those of immigrants and international students. This supports universal design approaches while maintaining attention to the specific linguistic profiles of different user groups. Limitations include the focus on computing professionals who may have higher literacy levels than the general DHH population, and the use of video demonstrations rather than working prototypes.
Tags: deaf · hard of hearing · DHH · automatic text simplification · reading assistance · social accessibility · natural language processing · workplace accessibility